Invited commentary: The use of sibship studies to detect familial confounding.

The authors discuss how the sibship design can be used to detect and control for familial confounding. Family-level confounding is especially problematic when estimating modest individual-level effects in the presence of familial confounders with large effects. This circumstance arises frequently in studies which relate indicators of fetal growth, such as birth weight, to outcomes that are strongly associated with parental socioeconomic status and genes. The study by Eriksen et al. in this issue of the Journal (Am J Epidemiol. 2010;172(5):530-536) uses the sibship design to capture the relation between birth weight, gestational age, and intelligence score among Norwegian males born as singletons at 37-41 completed weeks' gestation during 1967-1984. Their study illustrates how valuable the design can be in this kind of scenario. It also illustrates the potential complexity of sibship studies and the challenges they present for appropriate interpretation.

Associations of birth size, gestational age, and adult size with intellectual performance: evidence from a cohort of Norwegian men.

The influences of prenatal and postnatal growth on intellectual performance are unclear. We examined the associations of birth size and gestational age with intellectual performance and explored whether these associations were influenced by adult body size and social factors. In this nationwide cohort study, the records of 317,761 male infants registered in the Medical Birth Registry of Norway (1967-1979) were linked to the Norwegian Conscript Service (1984-1999). The variation in intelligence test score at age 18 due to birth weight and birth length was evaluated using absolute and standardized (z scores) values. Mean intelligence score increased by gestational age, birth weight, and birth length. However, a decline in intellectual performance was observed for gestational age >41 wk and birth weight >4500 g. There was a strong interaction on intellectual performance between birth size and gestational age (p < 0.0005). Adjusting for adult size strongly attenuated the association of birth size with intellectual performance. The overall R of intellectual performance explained by birth size was <1%; however, adding adult body size and social factors to the model increased R to 12%. In conclusion, the association of birth size with intellectual performance was weak, but still present after adjustment for adult body size and social factors.

Maternal epilepsy and offsprings' adult intelligence: a population-based study from Norway.

PURPOSE: We examined if children of mothers with epilepsy had normal intelligence, speculating that either epilepsy or its therapy might affect this parameter. METHODS: In a population-based cohort study in Norway, information on maternal epilepsy reported to the Medical Birth Registry of Norway, 1967-1979 was linked to information on 18-19-year-old men's intellectual ability and anthropometric measures ascertained by the Norwegian Conscripts Service, 1984-1999. The standardized intelligence test (IQ score) was reported as single-digit standard scores with values from 1 to 9. No individual information on antiepileptic drug therapy was available. RESULTS: Mean IQ score was lower in 1,207 conscripts whose mothers had epilepsy reported on the birth notification form, as compared with 316,554 conscripts of mothers without epilepsy; 4.8 (standard deviation 1.8) versus 5.2(1.8), p < 0.001, respectively. This difference remained after adjustment for maternal education, maternal age, birth order, marital status, year of birth, and weight and length at birth. When comparing men with and without reported maternal epilepsy, the odds ratio of having an IQ score < or = 3 was 1.6 (95% confidence interval: 1.4-1.8), which was unaffected by adjustment for confounding factors. Mean height among conscripts of mothers with and without epilepsy measured 178.6 and 179.9 cm, respectively, a difference of 1.3 cm (p < 0.001). CONCLUSIONS: We observed that almost 20 years after birth, maternal epilepsy was associated with reduced IQ score and also shorter adult height in male offspring. We do not know whether these findings will persist when epilepsy is better treated with the newer and safer generation of antiepileptic drugs.

Associations of birth defects with adult intellectual performance, disability and mortality: population-based cohort study.

Infants born with birth defects have poorer outcomes in terms of mortality and disability, but the long-term intellectual outcome in children with birth defects is generally unknown. We assessed the long-term associations of various birth defects with mortality and disability, and evaluated whether high mortality and disability were reflected in impaired intellectual performance at age 18. In this nationwide cohort study, records of 9,186 males with and 384,384 without birth defects, registered in the Medical Birth Registry of Norway (1967-1979) were linked to the National Conscript Service (1984-1999). Mortality and disability before military draft, and intelligence test score at conscription were the main outcome measures. Males with birth defects had a relative risk for disability of 6.0 compared with males without defects. Disability was low within categories of birth defects associated with low mortality, and high within defect categories associated with high mortality. The relative risk for not being drafted was highest if maternal educational level was low. Heart defects and cleft palate were the only subgroups in which intellectual performance was lower after adjustment for maternal education, maternal age, marital status and birth order. In particular, intellectual performance was not impaired among those with multiple compared with single defects. We conclude that for the majority of birth defect categories in the present birth cohort, our hypothesis that intellectual performance would be impaired was not confirmed. Thus, there seems to be little reason to fear an adverse intellectual outcome in non-disabled surviving infants with birth defects.

Size at birth and gestational age as predictors of adult height and weight.

BACKGROUND: Both birth length and birth weight are associated with height in adulthood and may have independent contributions to adult body size, but the effects of gestational age on these associations have not been fully evaluated. Our objective was to examine the independent contributions of gestational age, and of length and weight at birth, on adult (age 18 years) height and weight, with a special focus on the effects of being born preterm. METHODS: In this nationwide cohort study, records of 348,706 male infants included in the Medical Birth Registry of Norway (1967-1979) were linked to the Norwegian Conscripts Service (1984-1999). Complete follow-up information, including deaths, emigration, and disability pension, was obtained for 94%. We analyzed length and weight at birth using standardized (z-scores) values and stratified by gestational age. RESULTS: The positive association between birth length and adult height was stronger than between birth weight and adult weight (R = 7-9% compared with <0.1%, respectively). The strongest associations were seen among those born at gestational age 39 to 41 weeks. The effects of birth length on adult height, and of birth weight on adult weight, were considerably less among preterm births than among term births. Length and weight at birth each contributed independently to adult stature and body weight. The increase in adult weight per relative birth weight category was greatest for infants who were both heavy and long at birth. CONCLUSIONS: Birth length is perhaps a better predictor of adult height and weight than birth weight, and should be considered as a possible risk factor for adult morbidity and mortality.

Breech delivery and intelligence: a population-based study of 8,738 breech infants.

OBJECTIVE: Long-term intellectual performance in breech-presented infants may be negatively affected by vaginal delivery. We evaluated the effect of presentation at birth and delivery mode on intellectual performance at age 18 years in a nationwide population study. METHODS: We studied 8,738 male infants in breech and 384,832 males in cephalic presentation registered in the Medical Birth Registry of Norway, 1967-1979, and linked to data registered at the National Conscript Service, 1984-1999. Test scores of intelligence testing at conscription were presented as standard nine ("stanine") scores. Mean stanine scores and odds ratios of low score were computed and adjusted for birth order, maternal age, and education. RESULTS: Mean stanine score was slightly higher among breech-presented males than among cephalic-presented males (5.26 versus 5.22, P = .05), whereas after adjustment the difference disappeared (P = .3). Breech-presented infants had lower mean scores if delivered by cesarean compared with vaginal breech delivery (P = .03), and cephalic-presented males scored lower if their mothers had a cesarean delivery instead of a vaginal delivery (P < .001). Comparing cesarean and vaginal delivery in breech births, the odds ratio of having a stanine score less than or equal to 3 was 1.12 (95% confidence interval 0.92,1.36), after adjustment for confounding factors. CONCLUSION: Presentation at birth did not affect adult intellectual performance. Cesarean delivery of breech-presented infants did not improve adult intellectual performance when compared with a vaginal delivery. The excess perinatal hazards of breech-presented infants with a vaginal delivery were not reflected in adult intellectual performance.